1e1785e85SDave Hansenconfig SELECT_MEMORY_MODEL 2e1785e85SDave Hansen def_bool y 3e1785e85SDave Hansen depends on EXPERIMENTAL || ARCH_SELECT_MEMORY_MODEL 4e1785e85SDave Hansen 53a9da765SDave Hansenchoice 63a9da765SDave Hansen prompt "Memory model" 7e1785e85SDave Hansen depends on SELECT_MEMORY_MODEL 8e1785e85SDave Hansen default DISCONTIGMEM_MANUAL if ARCH_DISCONTIGMEM_DEFAULT 9d41dee36SAndy Whitcroft default SPARSEMEM_MANUAL if ARCH_SPARSEMEM_DEFAULT 10e1785e85SDave Hansen default FLATMEM_MANUAL 113a9da765SDave Hansen 12e1785e85SDave Hansenconfig FLATMEM_MANUAL 133a9da765SDave Hansen bool "Flat Memory" 14c898ec16SAnton Blanchard depends on !(ARCH_DISCONTIGMEM_ENABLE || ARCH_SPARSEMEM_ENABLE) || ARCH_FLATMEM_ENABLE 153a9da765SDave Hansen help 163a9da765SDave Hansen This option allows you to change some of the ways that 173a9da765SDave Hansen Linux manages its memory internally. Most users will 183a9da765SDave Hansen only have one option here: FLATMEM. This is normal 193a9da765SDave Hansen and a correct option. 203a9da765SDave Hansen 21d41dee36SAndy Whitcroft Some users of more advanced features like NUMA and 22d41dee36SAndy Whitcroft memory hotplug may have different options here. 23d41dee36SAndy Whitcroft DISCONTIGMEM is an more mature, better tested system, 24d41dee36SAndy Whitcroft but is incompatible with memory hotplug and may suffer 25d41dee36SAndy Whitcroft decreased performance over SPARSEMEM. If unsure between 26d41dee36SAndy Whitcroft "Sparse Memory" and "Discontiguous Memory", choose 27d41dee36SAndy Whitcroft "Discontiguous Memory". 28d41dee36SAndy Whitcroft 29d41dee36SAndy Whitcroft If unsure, choose this option (Flat Memory) over any other. 303a9da765SDave Hansen 31e1785e85SDave Hansenconfig DISCONTIGMEM_MANUAL 32f3519f91SDave Hansen bool "Discontiguous Memory" 333a9da765SDave Hansen depends on ARCH_DISCONTIGMEM_ENABLE 343a9da765SDave Hansen help 35785dcd44SDave Hansen This option provides enhanced support for discontiguous 36785dcd44SDave Hansen memory systems, over FLATMEM. These systems have holes 37785dcd44SDave Hansen in their physical address spaces, and this option provides 38785dcd44SDave Hansen more efficient handling of these holes. However, the vast 39785dcd44SDave Hansen majority of hardware has quite flat address spaces, and 40ad3d0a38SPhilipp Marek can have degraded performance from the extra overhead that 41785dcd44SDave Hansen this option imposes. 42785dcd44SDave Hansen 43785dcd44SDave Hansen Many NUMA configurations will have this as the only option. 44785dcd44SDave Hansen 453a9da765SDave Hansen If unsure, choose "Flat Memory" over this option. 463a9da765SDave Hansen 47d41dee36SAndy Whitcroftconfig SPARSEMEM_MANUAL 48d41dee36SAndy Whitcroft bool "Sparse Memory" 49d41dee36SAndy Whitcroft depends on ARCH_SPARSEMEM_ENABLE 50d41dee36SAndy Whitcroft help 51d41dee36SAndy Whitcroft This will be the only option for some systems, including 52d41dee36SAndy Whitcroft memory hotplug systems. This is normal. 53d41dee36SAndy Whitcroft 54d41dee36SAndy Whitcroft For many other systems, this will be an alternative to 55f3519f91SDave Hansen "Discontiguous Memory". This option provides some potential 56d41dee36SAndy Whitcroft performance benefits, along with decreased code complexity, 57d41dee36SAndy Whitcroft but it is newer, and more experimental. 58d41dee36SAndy Whitcroft 59d41dee36SAndy Whitcroft If unsure, choose "Discontiguous Memory" or "Flat Memory" 60d41dee36SAndy Whitcroft over this option. 61d41dee36SAndy Whitcroft 623a9da765SDave Hansenendchoice 633a9da765SDave Hansen 64e1785e85SDave Hansenconfig DISCONTIGMEM 65e1785e85SDave Hansen def_bool y 66e1785e85SDave Hansen depends on (!SELECT_MEMORY_MODEL && ARCH_DISCONTIGMEM_ENABLE) || DISCONTIGMEM_MANUAL 67e1785e85SDave Hansen 68d41dee36SAndy Whitcroftconfig SPARSEMEM 69d41dee36SAndy Whitcroft def_bool y 701a83e175SRussell King depends on (!SELECT_MEMORY_MODEL && ARCH_SPARSEMEM_ENABLE) || SPARSEMEM_MANUAL 71d41dee36SAndy Whitcroft 72e1785e85SDave Hansenconfig FLATMEM 73e1785e85SDave Hansen def_bool y 74d41dee36SAndy Whitcroft depends on (!DISCONTIGMEM && !SPARSEMEM) || FLATMEM_MANUAL 75d41dee36SAndy Whitcroft 76d41dee36SAndy Whitcroftconfig FLAT_NODE_MEM_MAP 77d41dee36SAndy Whitcroft def_bool y 78d41dee36SAndy Whitcroft depends on !SPARSEMEM 79e1785e85SDave Hansen 8093b7504eSDave Hansen# 8193b7504eSDave Hansen# Both the NUMA code and DISCONTIGMEM use arrays of pg_data_t's 8293b7504eSDave Hansen# to represent different areas of memory. This variable allows 8393b7504eSDave Hansen# those dependencies to exist individually. 8493b7504eSDave Hansen# 8593b7504eSDave Hansenconfig NEED_MULTIPLE_NODES 8693b7504eSDave Hansen def_bool y 8793b7504eSDave Hansen depends on DISCONTIGMEM || NUMA 88af705362SAndy Whitcroft 89af705362SAndy Whitcroftconfig HAVE_MEMORY_PRESENT 90af705362SAndy Whitcroft def_bool y 91d41dee36SAndy Whitcroft depends on ARCH_HAVE_MEMORY_PRESENT || SPARSEMEM 92802f192eSBob Picco 93802f192eSBob Picco# 943e347261SBob Picco# SPARSEMEM_EXTREME (which is the default) does some bootmem 953e347261SBob Picco# allocations when memory_present() is called. If this cannot 963e347261SBob Picco# be done on your architecture, select this option. However, 973e347261SBob Picco# statically allocating the mem_section[] array can potentially 983e347261SBob Picco# consume vast quantities of .bss, so be careful. 993e347261SBob Picco# 1003e347261SBob Picco# This option will also potentially produce smaller runtime code 1013e347261SBob Picco# with gcc 3.4 and later. 1023e347261SBob Picco# 1033e347261SBob Piccoconfig SPARSEMEM_STATIC 1049ba16087SJan Beulich bool 1053e347261SBob Picco 1063e347261SBob Picco# 10744c09201SMatt LaPlante# Architecture platforms which require a two level mem_section in SPARSEMEM 108802f192eSBob Picco# must select this option. This is usually for architecture platforms with 109802f192eSBob Picco# an extremely sparse physical address space. 110802f192eSBob Picco# 1113e347261SBob Piccoconfig SPARSEMEM_EXTREME 1123e347261SBob Picco def_bool y 1133e347261SBob Picco depends on SPARSEMEM && !SPARSEMEM_STATIC 1144c21e2f2SHugh Dickins 11529c71111SAndy Whitcroftconfig SPARSEMEM_VMEMMAP_ENABLE 1169ba16087SJan Beulich bool 11729c71111SAndy Whitcroft 1189bdac914SYinghai Luconfig SPARSEMEM_ALLOC_MEM_MAP_TOGETHER 1199bdac914SYinghai Lu def_bool y 1209bdac914SYinghai Lu depends on SPARSEMEM && X86_64 1219bdac914SYinghai Lu 12229c71111SAndy Whitcroftconfig SPARSEMEM_VMEMMAP 123a5ee6daaSGeoff Levand bool "Sparse Memory virtual memmap" 124a5ee6daaSGeoff Levand depends on SPARSEMEM && SPARSEMEM_VMEMMAP_ENABLE 125a5ee6daaSGeoff Levand default y 126a5ee6daaSGeoff Levand help 127a5ee6daaSGeoff Levand SPARSEMEM_VMEMMAP uses a virtually mapped memmap to optimise 128a5ee6daaSGeoff Levand pfn_to_page and page_to_pfn operations. This is the most 129a5ee6daaSGeoff Levand efficient option when sufficient kernel resources are available. 13029c71111SAndy Whitcroft 13195f72d1eSYinghai Luconfig HAVE_MEMBLOCK 13295f72d1eSYinghai Lu boolean 13395f72d1eSYinghai Lu 1347c0caeb8STejun Heoconfig HAVE_MEMBLOCK_NODE_MAP 1357c0caeb8STejun Heo boolean 1367c0caeb8STejun Heo 137c378ddd5STejun Heoconfig ARCH_DISCARD_MEMBLOCK 138c378ddd5STejun Heo boolean 139c378ddd5STejun Heo 14066616720SSam Ravnborgconfig NO_BOOTMEM 14166616720SSam Ravnborg boolean 14266616720SSam Ravnborg 143ee6f509cSMinchan Kimconfig MEMORY_ISOLATION 144ee6f509cSMinchan Kim boolean 145ee6f509cSMinchan Kim 14620b2f52bSLai Jiangshanconfig MOVABLE_NODE 14720b2f52bSLai Jiangshan boolean "Enable to assign a node which has only movable memory" 14820b2f52bSLai Jiangshan depends on HAVE_MEMBLOCK 14920b2f52bSLai Jiangshan depends on NO_BOOTMEM 15020b2f52bSLai Jiangshan depends on X86_64 15120b2f52bSLai Jiangshan depends on NUMA 15220b2f52bSLai Jiangshan default y 15320b2f52bSLai Jiangshan 1543947be19SDave Hansen# eventually, we can have this option just 'select SPARSEMEM' 1553947be19SDave Hansenconfig MEMORY_HOTPLUG 1563947be19SDave Hansen bool "Allow for memory hot-add" 157ee6f509cSMinchan Kim select MEMORY_ISOLATION 158ec69acbbSKeith Mannthey depends on SPARSEMEM || X86_64_ACPI_NUMA 1596ad696d2SAndi Kleen depends on HOTPLUG && ARCH_ENABLE_MEMORY_HOTPLUG 160ed84a07aSKumar Gala depends on (IA64 || X86 || PPC_BOOK3S_64 || SUPERH || S390) 1613947be19SDave Hansen 162ec69acbbSKeith Manntheyconfig MEMORY_HOTPLUG_SPARSE 163ec69acbbSKeith Mannthey def_bool y 164ec69acbbSKeith Mannthey depends on SPARSEMEM && MEMORY_HOTPLUG 165ec69acbbSKeith Mannthey 1660c0e6195SKAMEZAWA Hiroyukiconfig MEMORY_HOTREMOVE 1670c0e6195SKAMEZAWA Hiroyuki bool "Allow for memory hot remove" 1680c0e6195SKAMEZAWA Hiroyuki depends on MEMORY_HOTPLUG && ARCH_ENABLE_MEMORY_HOTREMOVE 1690c0e6195SKAMEZAWA Hiroyuki depends on MIGRATION 1700c0e6195SKAMEZAWA Hiroyuki 171e20b8ccaSChristoph Lameter# 172e20b8ccaSChristoph Lameter# If we have space for more page flags then we can enable additional 173e20b8ccaSChristoph Lameter# optimizations and functionality. 174e20b8ccaSChristoph Lameter# 175e20b8ccaSChristoph Lameter# Regular Sparsemem takes page flag bits for the sectionid if it does not 176e20b8ccaSChristoph Lameter# use a virtual memmap. Disable extended page flags for 32 bit platforms 177e20b8ccaSChristoph Lameter# that require the use of a sectionid in the page flags. 178e20b8ccaSChristoph Lameter# 179e20b8ccaSChristoph Lameterconfig PAGEFLAGS_EXTENDED 180e20b8ccaSChristoph Lameter def_bool y 181a269cca9SH. Peter Anvin depends on 64BIT || SPARSEMEM_VMEMMAP || !SPARSEMEM 182e20b8ccaSChristoph Lameter 1834c21e2f2SHugh Dickins# Heavily threaded applications may benefit from splitting the mm-wide 1844c21e2f2SHugh Dickins# page_table_lock, so that faults on different parts of the user address 1854c21e2f2SHugh Dickins# space can be handled with less contention: split it at this NR_CPUS. 1864c21e2f2SHugh Dickins# Default to 4 for wider testing, though 8 might be more appropriate. 1874c21e2f2SHugh Dickins# ARM's adjust_pte (unused if VIPT) depends on mm-wide page_table_lock. 1887b6ac9dfSHugh Dickins# PA-RISC 7xxx's spinlock_t would enlarge struct page from 32 to 44 bytes. 189a70caa8bSHugh Dickins# DEBUG_SPINLOCK and DEBUG_LOCK_ALLOC spinlock_t also enlarge struct page. 1904c21e2f2SHugh Dickins# 1914c21e2f2SHugh Dickinsconfig SPLIT_PTLOCK_CPUS 1924c21e2f2SHugh Dickins int 193a70caa8bSHugh Dickins default "999999" if ARM && !CPU_CACHE_VIPT 194a70caa8bSHugh Dickins default "999999" if PARISC && !PA20 195a70caa8bSHugh Dickins default "999999" if DEBUG_SPINLOCK || DEBUG_LOCK_ALLOC 1964c21e2f2SHugh Dickins default "4" 1977cbe34cfSChristoph Lameter 1987cbe34cfSChristoph Lameter# 19918468d93SRafael Aquini# support for memory balloon compaction 20018468d93SRafael Aquiniconfig BALLOON_COMPACTION 20118468d93SRafael Aquini bool "Allow for balloon memory compaction/migration" 20218468d93SRafael Aquini def_bool y 20318468d93SRafael Aquini depends on COMPACTION && VIRTIO_BALLOON 20418468d93SRafael Aquini help 20518468d93SRafael Aquini Memory fragmentation introduced by ballooning might reduce 20618468d93SRafael Aquini significantly the number of 2MB contiguous memory blocks that can be 20718468d93SRafael Aquini used within a guest, thus imposing performance penalties associated 20818468d93SRafael Aquini with the reduced number of transparent huge pages that could be used 20918468d93SRafael Aquini by the guest workload. Allowing the compaction & migration for memory 21018468d93SRafael Aquini pages enlisted as being part of memory balloon devices avoids the 21118468d93SRafael Aquini scenario aforementioned and helps improving memory defragmentation. 21218468d93SRafael Aquini 21318468d93SRafael Aquini# 214e9e96b39SMel Gorman# support for memory compaction 215e9e96b39SMel Gormanconfig COMPACTION 216e9e96b39SMel Gorman bool "Allow for memory compaction" 21705106e6aSRik van Riel def_bool y 218e9e96b39SMel Gorman select MIGRATION 21933a93877SAndrea Arcangeli depends on MMU 220e9e96b39SMel Gorman help 221e9e96b39SMel Gorman Allows the compaction of memory for the allocation of huge pages. 222e9e96b39SMel Gorman 223e9e96b39SMel Gorman# 2247cbe34cfSChristoph Lameter# support for page migration 2257cbe34cfSChristoph Lameter# 2267cbe34cfSChristoph Lameterconfig MIGRATION 227b20a3503SChristoph Lameter bool "Page migration" 2286c5240aeSChristoph Lameter def_bool y 22947118af0SMichal Nazarewicz depends on NUMA || ARCH_ENABLE_MEMORY_HOTREMOVE || COMPACTION || CMA 230b20a3503SChristoph Lameter help 231b20a3503SChristoph Lameter Allows the migration of the physical location of pages of processes 232e9e96b39SMel Gorman while the virtual addresses are not changed. This is useful in 233e9e96b39SMel Gorman two situations. The first is on NUMA systems to put pages nearer 234e9e96b39SMel Gorman to the processors accessing. The second is when allocating huge 235e9e96b39SMel Gorman pages as migration can relocate pages to satisfy a huge page 236e9e96b39SMel Gorman allocation instead of reclaiming. 2376550e07fSGreg Kroah-Hartman 238600715dcSJeremy Fitzhardingeconfig PHYS_ADDR_T_64BIT 239600715dcSJeremy Fitzhardinge def_bool 64BIT || ARCH_PHYS_ADDR_T_64BIT 240600715dcSJeremy Fitzhardinge 2414b51d669SChristoph Lameterconfig ZONE_DMA_FLAG 2424b51d669SChristoph Lameter int 2434b51d669SChristoph Lameter default "0" if !ZONE_DMA 2444b51d669SChristoph Lameter default "1" 2454b51d669SChristoph Lameter 2462a7326b5SChristoph Lameterconfig BOUNCE 2472a7326b5SChristoph Lameter def_bool y 2482a7326b5SChristoph Lameter depends on BLOCK && MMU && (ZONE_DMA || HIGHMEM) 2492a7326b5SChristoph Lameter 2506225e937SChristoph Lameterconfig NR_QUICK 2516225e937SChristoph Lameter int 2526225e937SChristoph Lameter depends on QUICKLIST 2530176bd3dSPaul Mundt default "2" if AVR32 2546225e937SChristoph Lameter default "1" 255f057eac0SStephen Rothwell 256f057eac0SStephen Rothwellconfig VIRT_TO_BUS 257f057eac0SStephen Rothwell def_bool y 258f057eac0SStephen Rothwell depends on !ARCH_NO_VIRT_TO_BUS 259cddb8a5cSAndrea Arcangeli 260cddb8a5cSAndrea Arcangeliconfig MMU_NOTIFIER 261cddb8a5cSAndrea Arcangeli bool 262fc4d5c29SDavid Howells 263f8af4da3SHugh Dickinsconfig KSM 264f8af4da3SHugh Dickins bool "Enable KSM for page merging" 265f8af4da3SHugh Dickins depends on MMU 266f8af4da3SHugh Dickins help 267f8af4da3SHugh Dickins Enable Kernel Samepage Merging: KSM periodically scans those areas 268f8af4da3SHugh Dickins of an application's address space that an app has advised may be 269f8af4da3SHugh Dickins mergeable. When it finds pages of identical content, it replaces 270d0f209f6SHugh Dickins the many instances by a single page with that content, so 271f8af4da3SHugh Dickins saving memory until one or another app needs to modify the content. 272f8af4da3SHugh Dickins Recommended for use with KVM, or with other duplicative applications. 273c73602adSHugh Dickins See Documentation/vm/ksm.txt for more information: KSM is inactive 274c73602adSHugh Dickins until a program has madvised that an area is MADV_MERGEABLE, and 275c73602adSHugh Dickins root has set /sys/kernel/mm/ksm/run to 1 (if CONFIG_SYSFS is set). 276f8af4da3SHugh Dickins 277e0a94c2aSChristoph Lameterconfig DEFAULT_MMAP_MIN_ADDR 278e0a94c2aSChristoph Lameter int "Low address space to protect from user allocation" 2796e141546SDavid Howells depends on MMU 280e0a94c2aSChristoph Lameter default 4096 281e0a94c2aSChristoph Lameter help 282e0a94c2aSChristoph Lameter This is the portion of low virtual memory which should be protected 283e0a94c2aSChristoph Lameter from userspace allocation. Keeping a user from writing to low pages 284e0a94c2aSChristoph Lameter can help reduce the impact of kernel NULL pointer bugs. 285e0a94c2aSChristoph Lameter 286e0a94c2aSChristoph Lameter For most ia64, ppc64 and x86 users with lots of address space 287e0a94c2aSChristoph Lameter a value of 65536 is reasonable and should cause no problems. 288e0a94c2aSChristoph Lameter On arm and other archs it should not be higher than 32768. 289788084abSEric Paris Programs which use vm86 functionality or have some need to map 290788084abSEric Paris this low address space will need CAP_SYS_RAWIO or disable this 291788084abSEric Paris protection by setting the value to 0. 292e0a94c2aSChristoph Lameter 293e0a94c2aSChristoph Lameter This value can be changed after boot using the 294e0a94c2aSChristoph Lameter /proc/sys/vm/mmap_min_addr tunable. 295e0a94c2aSChristoph Lameter 296d949f36fSLinus Torvaldsconfig ARCH_SUPPORTS_MEMORY_FAILURE 297d949f36fSLinus Torvalds bool 298e0a94c2aSChristoph Lameter 2996a46079cSAndi Kleenconfig MEMORY_FAILURE 3006a46079cSAndi Kleen depends on MMU 301d949f36fSLinus Torvalds depends on ARCH_SUPPORTS_MEMORY_FAILURE 3026a46079cSAndi Kleen bool "Enable recovery from hardware memory errors" 303ee6f509cSMinchan Kim select MEMORY_ISOLATION 3046a46079cSAndi Kleen help 3056a46079cSAndi Kleen Enables code to recover from some memory failures on systems 3066a46079cSAndi Kleen with MCA recovery. This allows a system to continue running 3076a46079cSAndi Kleen even when some of its memory has uncorrected errors. This requires 3086a46079cSAndi Kleen special hardware support and typically ECC memory. 3096a46079cSAndi Kleen 310cae681fcSAndi Kleenconfig HWPOISON_INJECT 311413f9efbSAndi Kleen tristate "HWPoison pages injector" 31227df5068SAndi Kleen depends on MEMORY_FAILURE && DEBUG_KERNEL && PROC_FS 313478c5ffcSWu Fengguang select PROC_PAGE_MONITOR 314cae681fcSAndi Kleen 315fc4d5c29SDavid Howellsconfig NOMMU_INITIAL_TRIM_EXCESS 316fc4d5c29SDavid Howells int "Turn on mmap() excess space trimming before booting" 317fc4d5c29SDavid Howells depends on !MMU 318fc4d5c29SDavid Howells default 1 319fc4d5c29SDavid Howells help 320fc4d5c29SDavid Howells The NOMMU mmap() frequently needs to allocate large contiguous chunks 321fc4d5c29SDavid Howells of memory on which to store mappings, but it can only ask the system 322fc4d5c29SDavid Howells allocator for chunks in 2^N*PAGE_SIZE amounts - which is frequently 323fc4d5c29SDavid Howells more than it requires. To deal with this, mmap() is able to trim off 324fc4d5c29SDavid Howells the excess and return it to the allocator. 325fc4d5c29SDavid Howells 326fc4d5c29SDavid Howells If trimming is enabled, the excess is trimmed off and returned to the 327fc4d5c29SDavid Howells system allocator, which can cause extra fragmentation, particularly 328fc4d5c29SDavid Howells if there are a lot of transient processes. 329fc4d5c29SDavid Howells 330fc4d5c29SDavid Howells If trimming is disabled, the excess is kept, but not used, which for 331fc4d5c29SDavid Howells long-term mappings means that the space is wasted. 332fc4d5c29SDavid Howells 333fc4d5c29SDavid Howells Trimming can be dynamically controlled through a sysctl option 334fc4d5c29SDavid Howells (/proc/sys/vm/nr_trim_pages) which specifies the minimum number of 335fc4d5c29SDavid Howells excess pages there must be before trimming should occur, or zero if 336fc4d5c29SDavid Howells no trimming is to occur. 337fc4d5c29SDavid Howells 338fc4d5c29SDavid Howells This option specifies the initial value of this option. The default 339fc4d5c29SDavid Howells of 1 says that all excess pages should be trimmed. 340fc4d5c29SDavid Howells 341fc4d5c29SDavid Howells See Documentation/nommu-mmap.txt for more information. 342bbddff05STejun Heo 3434c76d9d1SAndrea Arcangeliconfig TRANSPARENT_HUGEPAGE 34413ece886SAndrea Arcangeli bool "Transparent Hugepage Support" 34515626062SGerald Schaefer depends on HAVE_ARCH_TRANSPARENT_HUGEPAGE 3465d689240SAndrea Arcangeli select COMPACTION 3474c76d9d1SAndrea Arcangeli help 3484c76d9d1SAndrea Arcangeli Transparent Hugepages allows the kernel to use huge pages and 3494c76d9d1SAndrea Arcangeli huge tlb transparently to the applications whenever possible. 3504c76d9d1SAndrea Arcangeli This feature can improve computing performance to certain 3514c76d9d1SAndrea Arcangeli applications by speeding up page faults during memory 3524c76d9d1SAndrea Arcangeli allocation, by reducing the number of tlb misses and by speeding 3534c76d9d1SAndrea Arcangeli up the pagetable walking. 3544c76d9d1SAndrea Arcangeli 3554c76d9d1SAndrea Arcangeli If memory constrained on embedded, you may want to say N. 3564c76d9d1SAndrea Arcangeli 35713ece886SAndrea Arcangelichoice 35813ece886SAndrea Arcangeli prompt "Transparent Hugepage Support sysfs defaults" 35913ece886SAndrea Arcangeli depends on TRANSPARENT_HUGEPAGE 36013ece886SAndrea Arcangeli default TRANSPARENT_HUGEPAGE_ALWAYS 36113ece886SAndrea Arcangeli help 36213ece886SAndrea Arcangeli Selects the sysfs defaults for Transparent Hugepage Support. 36313ece886SAndrea Arcangeli 36413ece886SAndrea Arcangeli config TRANSPARENT_HUGEPAGE_ALWAYS 36513ece886SAndrea Arcangeli bool "always" 36613ece886SAndrea Arcangeli help 36713ece886SAndrea Arcangeli Enabling Transparent Hugepage always, can increase the 36813ece886SAndrea Arcangeli memory footprint of applications without a guaranteed 36913ece886SAndrea Arcangeli benefit but it will work automatically for all applications. 37013ece886SAndrea Arcangeli 37113ece886SAndrea Arcangeli config TRANSPARENT_HUGEPAGE_MADVISE 37213ece886SAndrea Arcangeli bool "madvise" 37313ece886SAndrea Arcangeli help 37413ece886SAndrea Arcangeli Enabling Transparent Hugepage madvise, will only provide a 37513ece886SAndrea Arcangeli performance improvement benefit to the applications using 37613ece886SAndrea Arcangeli madvise(MADV_HUGEPAGE) but it won't risk to increase the 37713ece886SAndrea Arcangeli memory footprint of applications without a guaranteed 37813ece886SAndrea Arcangeli benefit. 37913ece886SAndrea Arcangeliendchoice 38013ece886SAndrea Arcangeli 3815febcbe9SChristopher Yeohconfig CROSS_MEMORY_ATTACH 3825febcbe9SChristopher Yeoh bool "Cross Memory Support" 3835febcbe9SChristopher Yeoh depends on MMU 3845febcbe9SChristopher Yeoh default y 3855febcbe9SChristopher Yeoh help 3865febcbe9SChristopher Yeoh Enabling this option adds the system calls process_vm_readv and 3875febcbe9SChristopher Yeoh process_vm_writev which allow a process with the correct privileges 3885febcbe9SChristopher Yeoh to directly read from or write to to another process's address space. 3895febcbe9SChristopher Yeoh See the man page for more details. 3905febcbe9SChristopher Yeoh 391bbddff05STejun Heo# 392bbddff05STejun Heo# UP and nommu archs use km based percpu allocator 393bbddff05STejun Heo# 394bbddff05STejun Heoconfig NEED_PER_CPU_KM 395bbddff05STejun Heo depends on !SMP 396bbddff05STejun Heo bool 397bbddff05STejun Heo default y 398077b1f83SDan Magenheimer 399077b1f83SDan Magenheimerconfig CLEANCACHE 400077b1f83SDan Magenheimer bool "Enable cleancache driver to cache clean pages if tmem is present" 401077b1f83SDan Magenheimer default n 402077b1f83SDan Magenheimer help 403077b1f83SDan Magenheimer Cleancache can be thought of as a page-granularity victim cache 404077b1f83SDan Magenheimer for clean pages that the kernel's pageframe replacement algorithm 405077b1f83SDan Magenheimer (PFRA) would like to keep around, but can't since there isn't enough 406077b1f83SDan Magenheimer memory. So when the PFRA "evicts" a page, it first attempts to use 407140a1ef2SMichael Witten cleancache code to put the data contained in that page into 408077b1f83SDan Magenheimer "transcendent memory", memory that is not directly accessible or 409077b1f83SDan Magenheimer addressable by the kernel and is of unknown and possibly 410077b1f83SDan Magenheimer time-varying size. And when a cleancache-enabled 411077b1f83SDan Magenheimer filesystem wishes to access a page in a file on disk, it first 412077b1f83SDan Magenheimer checks cleancache to see if it already contains it; if it does, 413077b1f83SDan Magenheimer the page is copied into the kernel and a disk access is avoided. 414077b1f83SDan Magenheimer When a transcendent memory driver is available (such as zcache or 415077b1f83SDan Magenheimer Xen transcendent memory), a significant I/O reduction 416077b1f83SDan Magenheimer may be achieved. When none is available, all cleancache calls 417077b1f83SDan Magenheimer are reduced to a single pointer-compare-against-NULL resulting 418077b1f83SDan Magenheimer in a negligible performance hit. 419077b1f83SDan Magenheimer 420077b1f83SDan Magenheimer If unsure, say Y to enable cleancache 42127c6aec2SDan Magenheimer 42227c6aec2SDan Magenheimerconfig FRONTSWAP 42327c6aec2SDan Magenheimer bool "Enable frontswap to cache swap pages if tmem is present" 42427c6aec2SDan Magenheimer depends on SWAP 42527c6aec2SDan Magenheimer default n 42627c6aec2SDan Magenheimer help 42727c6aec2SDan Magenheimer Frontswap is so named because it can be thought of as the opposite 42827c6aec2SDan Magenheimer of a "backing" store for a swap device. The data is stored into 42927c6aec2SDan Magenheimer "transcendent memory", memory that is not directly accessible or 43027c6aec2SDan Magenheimer addressable by the kernel and is of unknown and possibly 43127c6aec2SDan Magenheimer time-varying size. When space in transcendent memory is available, 43227c6aec2SDan Magenheimer a significant swap I/O reduction may be achieved. When none is 43327c6aec2SDan Magenheimer available, all frontswap calls are reduced to a single pointer- 43427c6aec2SDan Magenheimer compare-against-NULL resulting in a negligible performance hit 43527c6aec2SDan Magenheimer and swap data is stored as normal on the matching swap device. 43627c6aec2SDan Magenheimer 43727c6aec2SDan Magenheimer If unsure, say Y to enable frontswap. 438